Routine activities of early life history stages of fishes occur in an intermediate hydrodynamic environment (as identified by Reynolds numbers), between a zone where drag is linearly dependent on velocity and resistive forces make large contributions to thrust, and a zone where inertial forces dominate except in the boundary layer immediately adjacent to the body surface. Sprint performance carries larvae into this latter zone; thus, locomotor activities important for survival of both larvae and adults occur in the same hydrodynamic environment and similar selective pressures would be expected to influence locomotor morphology of larvae and adults. The simplest framework for evaluating and interpreting development of larvae recognizes the parental form as the developmental terminus and uses adult forms as references to identify similarities and discrepancies in larva structure. Three measures of locomotor structure are used to examine changes during development: (a) the ratio of caudal peduncle depth to maximum body depth, which is small in thunniform fish; (b) a body shape factor, which is small in chaetodontiform fishes; and (c) a transient swimming thrust factor, which is large in esociform fish. Similarities in form are found in yolk-bearing larvae and are attributed to oxygen demand. Similarities in the pattern of development of larvae directly towards the parental form are found for fusiform species, except for delayed development of the caudal fin in tunas that probably is due to the hydrodynamic regime of the tail. Convergence with parents does not occur during the larva stage of deep-bodied, compressed species. This is attributed to high drag of compressed forms at low Reynolds numbers and diet differences between larvae and parents. Although this framework is successful in identifying and interpreting problems, comparative studies of form, kinematics, and performance during biologically important activities are essential.